linux-kernel - Re: [PATCH] stacktrace: print real address of trace entries

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20200904160736.GB21956@C02TD0UTHF1T.local>
Date:   Fri, 4 Sep 2020 17:07:36 +0100
From:   Mark Rutland <mark.rutland@....com>
To:     Haesung Kim <matia.kim@....com>
Cc:     jirislaby@...nel.org, akpm@...ux-foundation.org, mingo@...nel.org,
        geert@...ux-m68k.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] stacktrace: print real address of trace entries

On Fri, Sep 04, 2020 at 07:00:24PM +0900, Haesung Kim wrote:
> If function is marked as static and compiler decies to lnline function
> with or without inline keyword, the function has no symbol.
> We just know symbol located near the address of the inline function
> by %pS type that shows symbol and offset. But we don't know function
> name.

What exactly is output in this case today? Can't you get the real
addr/symbol from scripts/faddr2line?

> The real address let us extract the function name and location of
> source code by debugging tools such as addr2line. This is helpful to
> debug.

Not logging the address was a deliberate decision to minimize leakage of
the kernel's VA layout. This undermines that, and I don't think doing so
is a good idea.

If there is a problem with faddr2line we should work out how to improve
that.

Mark.

> 
> Signed-off-by: Haesung Kim <matia.kim@....com>
> ---
>  kernel/stacktrace.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c
> index 946f44a..b7168c5 100644
> --- a/kernel/stacktrace.c
> +++ b/kernel/stacktrace.c
> @@ -24,12 +24,15 @@ void stack_trace_print(const unsigned long *entries, unsigned int nr_entries,
>  		       int spaces)
>  {
>  	unsigned int i;
> +	unsigned long ip;
>  
>  	if (WARN_ON(!entries))
>  		return;
>  
>  	for (i = 0; i < nr_entries; i++)
> -		printk("%*c%pS\n", 1 + spaces, ' ', (void *)entries[i]);
> +		ip = entries[i];
> +		printk("%*c[<%px>] %pS\n",
> +			1 + spaces, ' ', (void *) ip, (void *) ip);
>  }
>  EXPORT_SYMBOL_GPL(stack_trace_print);
>  
> @@ -47,13 +50,15 @@ int stack_trace_snprint(char *buf, size_t size, const unsigned long *entries,
>  			unsigned int nr_entries, int spaces)
>  {
>  	unsigned int generated, i, total = 0;
> +	unsigned long ip;
>  
>  	if (WARN_ON(!entries))
>  		return 0;
>  
>  	for (i = 0; i < nr_entries && size; i++) {
> -		generated = snprintf(buf, size, "%*c%pS\n", 1 + spaces, ' ',
> -				     (void *)entries[i]);
> +		ip = entries[i];
> +		generated = snprintf(buf, size, "%*c[<%px>] %pS\n",
> +				     1 + spaces, ' ', (void *) ip, (void *) ip);
>  
>  		total += generated;
>  		if (generated >= size) {
> -- 
> 2.7.4
>